The lateral femoral notch sign and coronal lateral collateral ligament sign in magnetic resonance imaging failed to predict dynamic anterior tibial laxity

This study was a retrospective review performed in a tertiary medical center. After approval from the Institutional Review Board (IRB) at the National Cheng Kung University Hospital (A-ER-110-437), a retrospective review was performed on a cohort of all patients diagnosed with ACL tears at the authors’ institution between May 2020 and February 2022. Patients who had completed both an MRI and a GNRB arthrometer examination (Genourob, Laval, France) were included. Both exams were performed within 6 months after the injury. Patients with low resolution MRI, incomplete data from the chart review, previous ipsilateral knee ligament injuries or surgeries, and previous contralateral knee injury or surgeries were excluded. Relevant patient characteristics, including age, gender, and laterality, were collected and analyzed retrospectively.

The lateral femoral notch sign and the coronal LCL sign were evaluated from the patients’ MRI. The evaluations of these signs were performed separately by two operators (T-C H and C-K H) twice with a two-week interval between the evaluations. The intra- and inter-observer reliability were assessed by calculating the intra-class correlation coefficients (ICCs). ICC values less than 0.5, between 0.5 and 0.75, between 0.75 and 0.9, and greater than 0.90 are indicative of poor, moderate, good, and excellent reliability, respectively [8]. Imaging measurements were performed via a picture archiving and communication system (INFINITT PACS, INFINITT Healthcare Co. Ltd., South Korea).

The lateral femoral notch sign was identified on a T1-weighted sagittal MRI. The notch depth was measured as the distance between the deepest portion of the sulcus and the tangent line connecting the anterior and posterior edge of the lateral femoral notch [3] (Fig. 1). Based on previous studies [2, 3, 9, 10], a notch sign was considered to be positive if the notch depth was greater than 2.0 mm. A positive notch sign was further categorized into grade 1 (2.0 mm to 3.9 mm) and grade 2 (≥4.0 mm) [2]. We also measured the length of the notch sign, defined as the length of the tangent line between the anterior and posterior of the lateral femoral notch. The coronal T2 image or proton density image slices were reviewed on the MRI for the coronal LCL sign. The coronal LCL sign was considered positive when the LCL could be observed entirely from the femoral origin to the fibular insertion in a single coronal slice on the MRI image (Fig. 2) [6, 7].

The GNRB arthrometer (displacement in millimeters, with a 0.1 mm accuracy) was used to evaluate knee laxity by measuring the anterior tibial translation (Fig. 3). In accordance with the recommendations of the manufacturer of the GNRB (Genourob, Laval, France), the knee was placed in 20° flexion at a 0° rotation in a molded support in order to reproduce the typical Lachman test position [11, 12]. The data were automatically collected in the computer. The examinations were performed for both knees, and the side-to-side difference (SSD) between the injured knee and the healthy knee was calculated at each force level (89, 134, 150, and 200 N) for each subject with a 0.1 mm accuracy. For each applied force, the anterior tibial translation was measured simultaneously using the navigation system (3 measurements per knee). The displacement-force curve was created, and the slope of the curve was calculated. Slope 1 (S1) was defined as the slope of the curve ranging between 0 and 100 N, whereas slope 2 (S2) defined as the slope of the curve ranging between 100 N and the maximum force. The SSDs in S1 and S2 between the injured knee and healthy knee were calculated for the analysis.

Forty-six patients were initially included in the study. Four patients were excluded due to incomplete data, and 42 patients were finally enrolled for the analysis. The basic characteristics were summarized in (Table 1). Six patients (14.3%) had the positive lateral femoral notch sign, the average notch depth and notch width of which were 2.30 ± 0.2 mm and 14.7 ± 2.9 mm, respectively. All these positive notch signs were graded as grade 1 notch signs. Meanwhile, 14 patients (33.3%) in the study had positive coronal LCL signs. The mean age and gender distribution were the same for the aforementioned subgroups (Table 1). The interobserver repeatability and intraobserver repeatability of notch width, notch depth and LCL signs were summarized in Table 2.

The anterior tibial translation under different loads and slopes of the displacement-force curve from the GNRB for the different notch sign groups and the coronal LCL sign groups are summarized in Table 3. The SSDs of an anterior tibial translation under 134 N was the same for the positive notch sign group (3.8 ± 1.4 mm) and the negative notch group (4.0 ± 1.8 mm) (p = 0.875). Meanwhile, the SSD for S2 from the GNRB displacement-force curve was the same for the positive notch sign group (12.5 ± 8.9 mm/N) and the negative notch group (9.3 ± 6.8 mm/N) (p = 0.449). On the other hand, the SSDs in the anterior tibial translation under 134 N were the same for the positive coronal LCL sign group (3.9 ± 1.9 mm) and the negative coronal LCL sign group (4.0 ± 1.7 mm) (p = 0.722). The SSD of S2 from the GNRB displacement-force curve was the same for the positive coronal LCL sign group (10.7 ± 6.7 mm/N) and the negative coronal LCL sign group (9.2 ± 7.3 mm/N) (p = 0.348). The calculated effect sizes for these data ranged from 0.61 to 0.74. With a given α equal to 0.05, the post hoc achieved powers ranged from 48 to 61%.

There were no significant correlations between the anterior tibial translation SSD under different loads in the GNRB and the measured notch depth as well as notch width. Meanwhile, the notch depth and notch width were also not found to be correlated with the SSDs of S1 and S2 from the GNRB displacement-force curve. The detailed data are summarized in Table 4.